import numpy as np

#Creamos los arrays con los valores de Ltot y Rtot sobre los cuales vamos a obtener su error

lums = np.arange(22.375, 22.396, 0.001)
rads = np.arange(10.975, 10.99, 0.001)
error_mat = []

#Evaluamos el codigo original

for Rtot in rads:
    for Ltot in lums:
        
        %run Codigo_auxiliar.ipynb
        
        error_mat.append(Err_rel_total)
        
        print(Ltot, Rtot, Tcentral, Err_rel_total)

error_mat = np.array(error_mat).reshape(len(rads), len(lums))
error_mat

Probando_optimizacion_temperatura.ipynb:3: RuntimeWarning: invalid value encountered in double_scalars
  {
Probando_optimizacion_temperatura.ipynb:13: RuntimeWarning: invalid value encountered in double_scalars
  "import pandas as pd\n",
Probando_optimizacion_temperatura.ipynb:13: RuntimeWarning: invalid value encountered in double_scalars
  "import pandas as pd\n",

La temperatura central optima es 1.760
y se minimiza el error hasta el 0.311 por ciento

Probando_optimizacion_temperatura.ipynb:13: RuntimeWarning: invalid value encountered in double_scalars
  "import pandas as pd\n",

0.31138247078542486
22.375 10.975 1.76 0.31138247078542486

Probando_optimizacion_temperatura.ipynb:3: RuntimeWarning: invalid value encountered in double_scalars
  {
Probando_optimizacion_temperatura.ipynb:13: RuntimeWarning: invalid value encountered in double_scalars
  "import pandas as pd\n",
Probando_optimizacion_temperatura.ipynb:13: RuntimeWarning: invalid value encountered in double_scalars
  "import pandas as pd\n",

La temperatura central optima es 1.760
y se minimiza el error hasta el 0.310 por ciento

Probando_optimizacion_temperatura.ipynb:13: RuntimeWarning: invalid value encountered in double_scalars
  "import pandas as pd\n",

0.31025833181983237
22.376 10.975 1.76 0.31025833181983237

Probando_optimizacion_temperatura.ipynb:3: RuntimeWarning: invalid value encountered in double_scalars
  {
Probando_optimizacion_temperatura.ipynb:13: RuntimeWarning: invalid value encountered in double_scalars
  "import pandas as pd\n",
Probando_optimizacion_temperatura.ipynb:13: RuntimeWarning: invalid value encountered in double_scalars
  "import pandas as pd\n",

La temperatura central optima es 1.760
y se minimiza el error hasta el 0.309 por ciento

lums = np.arange(22.375, 22.396, 0.001)
rads = np.arange(10.975, 10.99, 0.001)

lums = lums.round(1)
rads = rads.round(1)

lums = np.array(lums).tolist()
rads = np.array(rads).tolist()
type(error_mat)

numpy.ndarray

import matplotlib.pyplot as plt
import numpy as np
import seaborn as sns

# Display matrix

font = {'family': 'serif',
        'color':  'black',
        'weight': 'normal',
        'size': 18,
        }

sns.set()

plt.figure(figsize=(40,21))
resultado = sns.heatmap(error_mat, cmap = 'hot', linewidths= 0.01, square=True)
#resultado = sns.heatmap(error_mat, cmap = 'coolwarm', annot = error_mat, linewidths= 0.5)

resultado.set_xticks(np.arange(len(lums)))
resultado.set_yticks(np.arange(len(rads)))

resultado.set_xticklabels(lums, fontdict=font)
resultado.set_yticklabels(rads, fontdict=font)

plt.title('Estudio del error relativo total', fontsize=40 )
plt.xlabel('luminosidad total ($L_{tot}$)', fontsize=35)
plt.ylabel('radio total ($R_{tot}$)', fontsize=35)

plt.savefig('mapacalorgrande.png')
plt.show(resultado)

error_minimo = error_mat.min()
error_minimo

0.25265756771430048

np.save('error_mat', error_mat)

Plot formatting ...¶

import numpy as np
import matplotlib as mpl
import matplotlib.pyplot as plt
%matplotlib inline

lums = np.arange(22.375, 22.396, 0.001).round(3)
rads = np.arange(10.975, 10.99, 0.001).round(3)
print(lums)
print(rads)

[ 22.375  22.376  22.377  22.378  22.379  22.38   22.381  22.382  22.383
  22.384  22.385  22.386  22.387  22.388  22.389  22.39   22.391  22.392
  22.393  22.394  22.395  22.396]
[ 10.975  10.976  10.977  10.978  10.979  10.98   10.981  10.982  10.983
  10.984  10.985  10.986  10.987  10.988  10.989  10.99 ]

import matplotlib.gridspec as gridspec
import matplotlib.ticker as ticker
from matplotlib.colors import LogNorm
plt.rcParams["font.family"] = "Times New Roman"

im = plt.imshow(error_mat, cmap='nipy_spectral');

import seaborn as sns

fig = plt.figure(figsize=(20,8))
ax = sns.heatmap(error_mat, cmap = 'nipy_spectral', linewidths= 0.01, cbar=False, square=True);

ax.set_xlabel('\nluminosidad total (L$_{tot}$)', fontsize=24);
ax.set_ylabel('radio total (R$_{tot}$)\n', fontsize=24);

lum_label = [];
for i in np.arange(0, len(lums), 1):
    if i%2 == 0:
        lum_label.append(lums[i])
    else:
        lum_label.append(' ');
        
rad_label = [];
for i in np.arange(0, len(rads), 1):
    if i%2 == 0:
        rad_label.append(rads[i])
    else:
        rad_label.append(' ');
    
ax.set_xticklabels(lum_label);
ax.set_yticklabels(rad_label);
ax.tick_params(labelsize = 18, axis = 'x', which = 'major', pad=10, direction='inout', length = 6, width = 1);
ax.tick_params(labelsize = 18, axis = 'y', which = 'major', pad=10, direction='inout', length = 6, width = 1);
    
plt.title('Estudio del error relativo total\n', fontsize=32);
plt.xticks(fontsize=18, rotation=0);
plt.yticks(fontsize=18, rotation=0);

from mpl_toolkits.axes_grid1 import make_axes_locatable
color_code = np.array([]);
divider = make_axes_locatable(ax);
cax = divider.append_axes('right', size='3%', pad=0.1);
cb = fig.colorbar(im, cax=cax);
#cb.set_label('relative error', fontsize=16, rotation=-90);
cb.solids.set_edgecolor('face');
cax.tick_params(labelsize=14, axis='y', which='major', pad=10, direction='inout', length=4, width=1);

#ax.annotate('Error mínimo', fontsize = 20,
            #xy=(4,13), xycoords='data',
            #xytext=(-180, -260), textcoords='offset points',
            #arrowprops=dict(arrowstyle="->"));


fig.tight_layout();
fig.savefig('rel_error_heatmap.png', bbox_inches='tight');
plt.show();

print(error_minimo)
print(lums[7])
print(rads[7])
error_mat[7,7]

0.252657567714
22.382
10.982

0.25265756771430048